- Email: [email protected]
ORIGINAL RESEARCH—OUTCOMES ASSESSMENT: Development and Validation of the Quality of Erection Questionnaire
Blackwell Publishing IncMalden, USAJSMJournal of Sexual Medicine1743-60952007 International Society for Sexual Medicine200742372381Original ArticlesQEQ Development and ValidationPorst et al.
372
ORIGINAL RESEARCH—OUTCOMES ASSESSMENT Development and Validation of the Quality of Erection Questionnaire Hartmut Porst, MD,* Claire Gilbert, MSc,† Suzanne Collins, BA,‡ Xiangning Huang, PhD,† Tara Symonds, PhD,† Vera Stecher, PhD,‡ and Kyle Hvidsten, MPH‡ *Private Practice of Urology/Andrology, Hamburg, Germany; †Pfizer Ltd., Sandwich, UK; ‡Pfizer Inc, New York, NY, USA DOI: 10.1111/j.1743-6109.2006.00422.x
ABSTRACT
Introduction. There are no psychometrically validated assessment tools designed to solely and specifically evaluate satisfaction with the quality of erections. Aim. To develop and psychometrically analyze the Quality of Erection Questionnaire (QEQ), a new patientreported measure developed to evaluate men’s satisfaction with the quality of their erections. Methods. The questionnaire was developed through in-depth qualitative interviews of men with erectile dysfunction (ED) in the United States and Australia. An exploratory methodology study was conducted on 65 men with ED. Subsequently, the psychometric properties were confirmed in a larger dataset of 558 men with ED from two combined clinical trials. Main Outcome Measures. Identification of potential redundancy or outliers in items (Pearson inter-item correlations); exploratory factor analysis (unrotated and varimax rotated); internal consistency (Cronbach’s alpha); convergent validity (Pearson correlation coefficients between the QEQ total score and domain scores of the International Index of Erectile Function); known-groups validity (ability of the QEQ scores to differentiate between ED severity groups); test–retest reliability (Pearson correlation coefficient). Results. The QEQ demonstrated excellent convergent and known-groups validity. Additional analysis demonstrated high internal consistency (Cronbach’s alpha, 0.92). Item analysis demonstrated a unidimensional structure and suggested that satisfaction with hardness may be the key driver for satisfaction with overall quality of erections (r = 0.8). The smaller exploratory study demonstrated good test–retest reliability (r = 0.82). Conclusions. The QEQ is a six-item, patient-reported measure with a unidimensional structure, which produces a total score that may be transformed to a 0–100 scale. Psychometric analysis confirmed reliability and validity of the QEQ, which solely and specifically evaluates satisfaction of men with the quality of their erections. The QEQ is a potentially useful measure for monitoring and evaluating treatment in those who are bothered by, or concerned about, their erectile function. Porst H, Gilbert C, Collins S, Huang X, Symonds T, Stecher V, and Hvidsten K. Development and validation of the Quality of Erection Questionnaire. J Sex Med 2007;4:372–381. Key Words. Erectile Dysfunction; Sildenafil; Quality of Erection Questionnaire; PDE 5 Inhibitors
Introduction
P
opulation-based data indicate that erectile dysfunction (ED) affects a substantial proportion of men worldwide [1]. In a cross-national survey, the prevalence of men aged 20–75 years J Sex Med 2007;4:372–381
with ED was 19–25% in the United States, Germany, Italy, and the UK, and was 12–13% in France and Spain [2]. The American Urological Association, the European Association of Urology, and the 2nd International Consultation on Sexual Medicine recommend oral phosphodiesterase type © 2006 International Society for Sexual Medicine
373
QEQ Development and Validation 5 inhibitors (PDE5Is; e.g., sildenafil citrate [Viagra®, Pfizer Inc], tadalafil [Cialis®, Lilly/ICOS], and vardenafil [Levitra®, Bayer/GSK]) as first-line oral therapy for ED, unless contraindicated [3–5]. The National Institute of Health Consensus Conference recommended improved “quantitative and qualitative classification of erectile dysfunction” and “standardization of patient assessment and treatment outcome” [6]. In response, several tools have been developed. The International Index of Erectile Function (IIEF), a 15-item patient-reported questionnaire accepted by regulatory agencies and scientific journals [7], is a standard tool for evaluating the efficacy of therapeutic intervention in men with ED in clinical trials [8– 10]. The IIEF Erectile Function domain is used to diagnose and classify severity of ED and to measure treatment effects in clinical research [9,11,12]. The abridged five-item version of the IIEF (IIEF-5), also known as the Sexual Health Inventory for Men (SHIM), was developed to diagnose the presence and severity of ED [13]. The inclusion of satisfaction domains in the IIEF (Intercourse Satisfaction and Overall Satisfaction) and of a single satisfaction question in the IIEF-5/SHIM (Q5: When you attempted sexual intercourse, how often was it satisfactory for you?) reflects the understanding that concern or bother of the patient is important to the evaluation and clinical management of ED. Concern or bother is essential to motivate the patient to seek and maintain treatment. Concern or bother about ED declines with increasing age of the patient [14,15]. In the previously mentioned cross-national survey, lack of concern or bother is inherent in some of the barriers cited most frequently by treatment-naive men as preventing their obtaining treatment, for example, the belief that ED is a normal part of aging (44%) and regarding ED as unimportant (17%) [2]. An international survey of more than 3,500 men with ED established that quality of erections, and specifically the capability to enable hard erections, is the primary attribute sought in a treatment for ED [16]. Although quality of erections is among the concepts included in the IIEF, its Erectile Function domain, and the IIEF-5/SHIM, hardness per se is addressed by only a single item in these questionnaires (Q2: When you had erections with sexual stimulation, how often were your erections hard enough for penetration?) and satisfaction with quality of erections is not addressed at all.
This report documents the development and psychometric validation of the Quality of Erection Questionnaire (QEQ), a patient-reported outcome for solely and specifically evaluating and monitoring satisfaction with the quality of erections in men who are concerned or bothered by their erectile function. Methods
Questionnaire Development The initial conceptual framework and the pool of items were developed through in-depth qualitative interviews of 35 men with ED, aged 18–60 years. The interviews were conducted in the United States (N = 20) and Australia (N = 15). Participants within the United States either were identified through lists of patients involved in clinical trials of ED medication or were respondents to advertisements in local newspapers. Participants within Australia were identified through routine outpatient and general practitioner consultation. All participants gave informed consent. They were required to have experience with some kind of treatment for their ED within the last 6 months and to be currently involved in a sexual relationship. ED was confirmed by a score of 21 or less (indicating at least mild-tomoderate severity) on the IIEF Erectile Function domain. The interviews began with a semi-structured open discussion to obtain perspectives of each individual on factors that relate to the quality of erections. This was followed by a request for feedback on a list of potential items that were suggested by a review of the literature. The semistructured discussions were analyzed to develop patient-derived concepts. Based on analysis of the interviews, a seven-item draft questionnaire was developed. Psychometric Evaluation Studies Exploratory Study. In a small exploratory study, the structure of the draft questionnaire was assessed, test–retest reliability was determined, and preliminary validation data were obtained. This study included 65 men with ED who were recruited during routine clinical consultation. Inclusion criteria were age 18–65 years, being in a sexual relationship, and receipt of treatment for ED for 2 months within the 6 months before enrollment. The draft questionnaire was comJ Sex Med 2007;4:372–381
374 pleted twice, with a minimum 10-day interval between completions. Clinical Trials Study. For the primary analysis of the psychometric properties of the questionnaire, data were obtained from the combined populations of two clinical trials, both assessing the efficacy and safety of an investigational new-generation PDE5I for the treatment of ED. This provided a larger population (N = 558), lending greater strength to the results. These two trials used identical inclusion/exclusion criteria, including recent successful treatment with a PDE5I (a washout period ensured that they were untreated at baseline). Baseline characteristics were compared statistically (Wilcoxon rank sum test for ordinal variables [i.e., age, duration of ED, and scores for the draft questionnaire and the IIEF] and chi-square for categorical variables [i.e., smoking and alcohol use]) to ensure the appropriateness of combining the data into a single, large dataset. The psychometric analyses were based on baseline data from the draft QEQ and IIEF. The trials were performed in compliance with relevant laws and institutional guidelines. Patients provided written informed consent before participation.
Psychometric Analyses Pearson inter-item correlations were used to identify any potential redundancy or outliers in items. Specifically, items that correlated highly (correlation coefficient, r > 0.8; e.g., coherence) were considered further to explore whether both items were needed (redundancy), and items that correlate poorly (r < 0.3) were considered further to examine whether they were covering the same concepts as other questions (outliers). The questionnaire’s underlying structure was evaluated using exploratory factor analysis (unrotated and varimax rotated); a loading of >0.5 delineated unidimensionality of items. Internal consistency was assessed by calculating Cronbach’s alpha. Convergent validity was evaluated by determining parametric correlation (Pearson correlation coefficients) between the QEQ total score and the IIEF domain scores. Known-groups validity was also evaluated by determining the ability of the QEQ scores (total and items) to differentiate between ED severity groups as defined by scores on the Erectile Function domain of the IIEF: 26– 30 (no ED), 22–25 (mild ED), 17–21 (mild-tomoderate ED), 11–16 (moderate ED), and 6–10 (severe ED). Because the QEQ scores were not J Sex Med 2007;4:372–381
Porst et al. normally distributed, nonparametric one-way analysis of variance was applied. Test–retest reliability was assessed across a 2week interval using Pearson correlation coefficient (r) ≥0.8. Because the protocols of the clinical trials did not provide a period of no expected change in ED, test–retest (unlike the other results reported herein) was based on data from the exploratory study. Results
The exploratory study assessing test–retest reliability included 65 men aged 22–75 years (mean ± SD, 54.7 ± 10.6 years) with ED duration of 2 months to 20 years (6.1 ± 4.4 years). ED medication had first been used from 1 month to 18 years previously (4.3 ± 3.8 years), 57 of 63 respondents (91%) indicated that they were currently using the prescribed medication, and usage was from 0 to 15 times per month (5.1 ± 3.4). The most commonly prescribed medications for ED were sildenafil (N = 54) and tadalafil (N = 44). The clinical trials enrolled 594 men aged 23– 65 years (53.8 ± 8.1 years) with ED duration of 0.5–35 years (6.2 ± 4.9 years), 558 of whom completed the draft questionnaire and composed the main study population. Of these, erectile function was categorized according to the IIEF Erectile Function domain in 546, including 214 (39%) with severe ED, 174 (32%) with moderate ED, 115 (21%) with mild-to-moderate ED, 36 (7%) with mild ED, and seven (1%) with no ED (who were included in the clinical trials based on a clinical diagnosis of ED). The mean IIEF Erectile Function domain score was 12.6 (out of 30). Between the two clinical trial populations that composed the main study population (N = 223 and N = 335, respectively), statistically significant differences were found in reported alcohol use (62% [119/191] vs. 76% [208/272] of respondents), the mean duration of ED (5.3 [range 0.5– 31] vs. 6.9 [range 0.7–35] years), and mean IIEF Orgasmic Function domain scores (6.0 vs. 5.5 out of 10) at baseline. These differences were not expected to affect the relationships between the QEQ and the other endpoints.
Quality of Completion There was a low rate of missing items—only eight of 558 had at least one missing item. There were no observed floor or ceiling effects, defined as an endorsement >50% for any response option. The highest endorsement frequencies were at the
375
QEQ Development and Validation Table 1
Parametric correlations between item scores on the draft questionnaire Pearson’s correlation coefficient (r value), N = 550* Frequency items
Satisfaction items
Draft questionnaire items†
Hard enough
Loss of erection
Keep erection
Time to hard enough
Time erect
Hardness
Overall quality
Hard enough Loss of erection Keep erection Time to hard enough Time erect Hardness Overall quality
1.00 0.38 0.58 0.63 0.55 0.62 0.61
— 1.00 0.44 0.26 0.40 0.36 0.40
— — 1.00 0.57 0.74 0.65 0.72
— — — 1.00 0.65 0.69 0.64
— — — — 1.00 0.71 0.75
— — — — — 1.00 0.82
— — — — — — 1.00
*Bold indicates outcomes that are highly correlated (>0.8) or poorly correlated (<0.3). † Draft questionnaire items were as follows: Frequency that: You had erections hard enough for penetration of your partner. You experienced the loss of your erection within a few minutes of penetration of your partner. Satisfaction with: Your ability to keep your erection to completion of sexual intercourse. The length of time (from when you started sexual activity) until your erection was hard enough to participate in sexual intercourse. The length of time you were able to be erect during intercourse. The hardness of your erection. The overall quality of your erection.
lower end of the score scale, indicating lower erection quality and satisfaction. This was expected from a population of men whose ED is untreated. There were no endorsements at the higher end of the score scale, suggesting room to capture any improvements in erection quality and in satisfaction.
Questionnaire Structure Inter-item correlations were relatively low between the “loss of erection” item (i.e., the frequency that “You experienced the loss of your erection within a few minutes of penetration of your partner”) and all other items (Table 1). This suggested that the “loss of erection” item may be an outlier, which needed to be evaluated further to determine its relationship to the other items and to the overall questionnaire structure. The strong correlation (r > 0.8) between the items questioning satisfaction with “The hardness of your erection” and “The overall quality of your erection” suggested potential redundancy. It could be that hardness is the key driver for overall satisfaction. All other items correlated moderately with each other (r = 0.55–0.75), with no obvious clusters of highly correlated items, suggesting a lack of redundancy and that the questionnaire is unidimensional. Factor analyses were used to identify the most appropriate factor structure for the questionnaire (Table 2). The results, quantified by eigenvalues, suggested a one-factor solution in which all seven items measure a single factor, accounting for 65%
of the variance. All other factors accounted for 12% or less of the variance. All items loaded onto the factor with a loading >0.5, which supports the unidimensionality of the questionnaire. However, the “loss of erection” item had a loading of 0.53, which was markedly lower than those of the other items (0.78–0.89). This suggests that the “loss of erection” item may not be a strong item in the overall factor structure. Further support for this conclusion came from this item’s low communality estimate (h2 = 0.28) compared with those of the other items (h2 = 0.61–0.79). This suggested that the “loss of erection” item (unlike the other items) is not strongly influenced by the one factor identified. Thus, the results of the item and factor analysis gave strong support for the questionnaire evaluating a unidimensional construct, with the removal of the “loss of erection” item. The resultant sixitem questionnaire, hereafter referred to as the QEQ, was scored by a single total score, calculated as the sum of the scores of all six items transformed onto a 0–100 scale (higher score = better quality).1 On this scale, the mean QEQ total score for the main study population was 25.6 ± 23.4. The transformed QEQ total score correlated positively with each item, with
1
The raw score may be used informally (e.g., in the clinical practice setting). However, Pfizer clinical trial data are reported as transformed scores, and transformed scores are recommended for use in clinical trials.
J Sex Med 2007;4:372–381
376
Porst et al.
Table 2 Factor analysis of the draft questionnaire: rotated and unrotated to identify the most appropriate factor structure, item loading to support unidimensionality, and communality estimates of the items Eigenvalues of the correlation matrix, N = 550 Eigenvalue Difference
Proportion
Cumulative
4.56 0.81 0.52 0.37 0.33 0.23 0.18
0.65 0.12 0.07 0.05 0.05 0.03 0.03
0.65 0.77 0.85 0.90 0.94 0.98 1.0
3.75 0.29 0.16 0.04 0.11 0.05 —
Factor pattern and communality estimates, N = 550
Factor pattern: item loading onto factor 1 (rotated and unrotated) Final communality estimates (h 2): factor 1 (rotated and unrotated): Total = 4.6
Frequency items*
Satisfaction items*
Hard enough
Loss of erection
Keep erection
Time to hard enough
Time erect
Hardness
Overall quality
0.78
0.53
0.85
0.80
0.87
0.88
0.89
0.61
0.28
0.72
0.64
0.75
0.77
0.79
*Draft questionnaire items were as follows: Frequency that: You had erections hard enough for penetration of your partner. You experienced the loss of your erection within a few minutes of penetration of your partner. Satisfaction with: Your ability to keep your erection to completion of sexual intercourse. The length of time (from when you started sexual activity) until your erection was hard enough to participate in sexual intercourse. The length of time you were able to be erect during intercourse. The hardness of your erection. The overall quality of your erection.
Spearman correlation coefficients ranging from 0.82 to 0.88 (Table 3). Given the high factor loading of the item about satisfaction with “the overall quality of your erection,” the factor analysis was rerun on the final QEQ without this item to ensure that this one item was not itself determining the single factor structure. The factor structure did not change with this reanalysis. The results, quantified by eigenvalues, suggested a one-factor solution in Table 3 Spearman’s nonparametric correlations between the QEQ total score (representing the unidimensionality of the questionnaire) and the scores for each item QEQ items†
Correlations (r values)*
Item Item Item Item Item Item
0.82 0.83 0.85 0.86 0.88 0.87
1: hard enough 2: keep erection 3: time to hard enough 4: time erect 5: hardness 6: overall quality
*Significant for all correlation coefficients, with P < 0.0001 (N = 550). † QEQ items were as follows: Frequency that: Item 1: You had erections hard enough for penetration of your partner. Satisfaction with: Item 2: Your ability to keep your erection to completion of sexual intercourse. Item 3: The length of time (from when you started sexual activity) until your erection was hard enough to participate in sexual intercourse. Item 4: The length of time you were able to be erect during intercourse. Item 5: The hardness of your erection. Item 6: The overall quality of your erection. QEQ = Quality of Erection Questionnaire.
J Sex Med 2007;4:372–381
which the remaining items measured a single factor, accounting for 71% of the variance. All other factors accounted for 10% or less of the variance. All items loaded onto the factor with a loading >0.79, which supports the unidimensionality of the questionnaire. A unidimensional structure was therefore accepted for the QEQ, using all six items of the final measure.
Reliability Reliability was demonstrated by internal consistency and by test–retest analysis. The interval between the first and second completion of the questionnaire for test–retest ranged from 10 to 50 days (mean ± SD, 19 ± 7.7 days). Fifty-four pairs of questionnaires were available for analysis by Pearson correlation coefficient. The QEQ had excellent internal consistency (Cronbach’s alpha = 0.92) and demonstrated good retest reliability (correlation coefficient, 0.82). Validity Because convergent validity demonstrates that endpoints are closely related (although not identical), it was expected that the strongest convergent relationships of the QEQ total score would be with the IIEF Overall Satisfaction domain and Erectile Function domain. The results confirmed
377
QEQ Development and Validation Table 4
Correlations between transformed QEQ scores and IIEF domain scores Pearson’s correlation coefficient (r value)* IIEF domain
QEQ†
Erectile function
Orgasmic function
Sexual desire
Intercourse satisfaction
Overall satisfaction
Total score Item 1. hard enough Item 2. keep erection Item 3. time to hard enough Item 4. time erect Item 5. hardness Item 6. overall quality
0.67 0.73 0.55 0.55 0.52 0.54 0.53
0.37 0.37 0.30 0.31 0.27 0.29 0.29
0.13 0.11 0.08 0.17 0.08 0.13 0.10
0.47 0.51 0.36 0.40 0.36 0.38 0.36
0.58 0.50 0.49 0.46 0.48 0.50 0.51
*P < 0.0001 except for correlations between the Sexual Desire domain and the QEQ total score, item 1, and item 5 (P < 0.01); item 6 (P < 0.05); and items 2 and 4 (P > 0.05, not statistically significant at the 0.05 level). † QEQ items were as follows: Frequency that: Item 1: You had erections hard enough for penetration of your partner. Satisfaction with: Item 2: Your ability to keep your erection to completion of sexual intercourse. Item 3: The length of time (from when you started sexual activity) until your erection was hard enough to participate in sexual intercourse. Item 4: The length of time you were able to be erect during intercourse. Item 5: The hardness of your erection. Item 6: The overall quality of your erection. QEQ = Quality of Erection Questionnaire; IIEF = International Index of Erectile Function.
this expectation, demonstrating correlations in the range expected for convergent validity for these domains, as well as for the Intercourse Satisfaction domain (r = 0.47–0.67) (Table 4). In the known-groups validity analysis, mean ± SD QEQ total scores followed the expected direction, being highest in the men categorized as having no ED and decreasing with each subsequent category (Table 5). The same relationship was seen for each individual QEQ item (Figure 1). Statistical analysis showed that these differences were significant (P < 0.0001 for an F value of 109.64 > F0.0001(4,541)). These results suggest that the QEQ is able to differentiate between all groups categorized by ED severity (P < 0.0001, with the exception of no ED vs. mild ED [P = 0.0031]), thereby demonstrating good known-groups validity.
Table 5 Nonparametric post-hoc analysis: difference in QEQ total score by ED severity QEQ total score ED severity*
Mean ± SD
Range
No ED (N = 7) Mild ED (N = 36) Mild-to-moderate ED (N = 115) Moderate ED (N = 174) Severe ED (N = 214)
79.8 ± 11.1 59.7 ± 16.4 39.9 ± 14.9 25.5 ± 17.5 10.5 ± 19.1
62.5–91.7 33.3–100 4.2–79.2 0–87.5 0–100
*ED severity was categorized by score on the Erectile Function domain of the International Index of Erectile Function. ED = erectile dysfunction; QEQ = Quality of Erection Questionnaire.
Discussion
The results show that the QEQ has good psychometric properties. The original seven items were reduced to a six-item unidimensional questionnaire. All items demonstrated good internal consistency with a strong factor structure and good test–retest reliability. The detailed analyses of the items and factor structure of the measure suggest that hardness is the key driver for overall satisfaction with the quality of erections, confirming the results of the international survey of more than 3,500 men with ED [16]. The concept that hardness is a key driver of overall satisfaction with the quality of erections is also confirmed by the qualitative work supporting the development of the questionnaire, in which some men reported overlap between the two items. Although this and the high level of correlation found between these items could suggest potential item redundancy, other men indicated as part of the initial qualitative work that overall quality for them reflected not only hardness but also other factors such as the additional QEQ concepts and emotional elements. The overall quality of erections item gives the respondent the chance to capture all the elements that are important to them in their overall perception of quality of their erections, whether that be primarily a matter of hardness for some or an amalgamation of several factors for others. Convergent and known-groups validity was demonstrated against the IIEF, a documented J Sex Med 2007;4:372–381
378
Porst et al. No ED (N=7) Mild ED (N=36) Mild-to-Moderate ED (N=116) Moderate ED (N=176) Severe ED (N=217–218)
QEQ Score by IIEF ED Severity Group
5
4
3
2
1
0 Item 1: Hard Enough
Item 2: Keep Erection
Item 3: Time to Hard Enough
Item 4: Time Erect
Item 5: Hardness
standard for evaluating ED. With the exception of the inability to differentiate between the two least impaired categories (“no ED” and “mild ED”), the QEQ was able to differentiate between all severity groups of ED categorized by IIEF Erectile Function domain scores. The inability to differentiate between “no ED” and “mild ED” may be the result of the small sample size of these groups (N = 36 and N = 7, respectively). Although the ability of the QEQ to differentiate between functional men and men with compromised erectile function warrants further research, the clear differentiation among the ED severity groups in the current study suggests that differentiation from functional men is likely given sufficient sample size. Although the IIEF Erectile Function domain and the QEQ total score are highly correlated (r = 0.67), the IIEF Erectile Function domain was designed to assess erectile function. The QEQ provides a further level of information by assessing satisfaction with the quality of the erections that were attained and maintained. The QEQ addresses the unmet need for a measure to solely and specifically assess the key concept of satisfaction with quality of erections from the individual male/ patient perspective. Quality of erections is composed of several aspects, such as the degree of hardness, the functionality of hardness, the time to onset of hardness, and the duration of hardness. Most previous methods for assessing quality of J Sex Med 2007;4:372–381
Item 6: Overall Quality
Figure 1 Known-groups discrimination: QEQ item score according to ED severity. ED severity was categorized by score on the Erectile Function domain of the IIEF. Item 1: (frequency that) You had erections hard enough for penetration of your partner; item 2: (satisfaction with) Your ability to keep your erection to completion of sexual intercourse; item 3: (satisfaction with) The length of time (from when you started sexual activity) until your erection was hard enough to participate in sexual intercourse; item 4: (satisfaction with) The length of time you were able to be erect during intercourse; item 5: (satisfaction with) The hardness of your erection; item 6: (satisfaction with) The overall quality of your erection. ED = erectile dysfunction; IIEF = International Index of Erectile Function; QEQ = Quality of Erection Questionnaire.
erections only addressed individual aspects: the unvalidated Erection Hardness Score, which allows a man to subjectively classify hardness on a simple 4-point scale at the time of erection; individual unvalidated items from validated questionnaires (e.g., IIEF Q2 [frequency of erection hard enough for penetration]); and individual questions with dichotomous (yes/no) response options, such as a global efficacy question or the questions of the Sexual Encounter Profile (“Were you able to insert your penis into your partner’s vagina?” and “Did your erection last long enough for you to have successful intercourse?”) [17]. In contrast, two recently developed questionnaires are much broader [18–20], The 15-item Erection Quality Scale (EQS) addresses several aspects of erection quality (six items) and also qualifies confidence in erection quality (three items) and satisfaction with erection quality (four items); the other two items address pleasurable feelings [18,20]. The 13-item Treatment Satisfaction Scale (TSS) addresses erection quality (one item) and also qualifies confidence in erection quality (two items) and satisfaction with erection quality (three items); the seven other items address pleasurable feelings and ED treatment [19]. Although both the EQS and the TSS are valuable tools, only the QEQ focuses solely and specifically on satisfaction with the quality of erections, which is a key quality-of-life measure from the individual male/patient perspective because lack of satisfaction is a prerequisite to
379
QEQ Development and Validation concern or bother, the drivers of treatment-seeking behavior. The quality of the data obtained suggests that the QEQ is acceptable to the ED population. This is to be expected because the items were derived from qualitative work with this population. Further analyses are being or have been conducted on the QEQ, including responsiveness/sensitivity to change, confirming the validity of the QEQ. A recent open-label, noncomparative, multicenter trial of sildenafil (50 or 100 mg as needed for 10 weeks) in men with ED who were in a stable, sexual relationship for at least 6 months and had limited previous PDE5I use (≤6 doses ever and no doses more recently than the previous 4 weeks) found that the mean ± SD transformed QEQ total score tripled from 22.0 ± 21.1 to 69.9 ± 35.9 (P < 0.0001) and the change correlated positively with change scores on the IIEF domains (r = 0.29– 0.86), change scores on measures of emotional well-being from the Self-Esteem And Relationship questionnaire (SEAR, r = 0.37–0.78), and the end-of-treatment score on the Erectile Dysfunction Inventory of Treatment Satisfaction (EDITS, r = 0.71) [21]. The SEAR questionnaire is a validated 14-item questionnaire consisting of a Sexual Relationship domain, a Confidence domain (i.e., Self-Esteem and Overall Relationship subscales), and an Overall score; it has been shown in controlled clinical trials to be responsive to effective treatment of ED with sildenafil [22,23]. The EDITS is an 11-item validated questionnaire that evaluates satisfaction with treatment for ED [24]. Thus, a relationship has been established between the broader psychosocial status of men with ED and satisfaction with the quality of their erections. The QEQ is a brief (six-item) and easy-to-use measure that solely and specifically evaluates satisfaction with the quality of erections from the patient’s perspective. This perspective enables the evaluation of treatment against the perceived need of the patient, and thus the QEQ is a potentially useful measure for use in clinical trials research or in the clinical practice setting. Details of the QEQ are given in the Appendixes. Acknowledgments
This study was sponsored by Pfizer Inc. Editorial support was provided by Deborah M. Campoli-Richards, RPh, of Complete Healthcare Communications, Inc., and was funded by Pfizer Inc. Correspondence Private Practice
Author: Hartmut Porst, MD, of Urology/Andrology, Neuer
Jungfernstieg 6a, Hamburg, 20354. Germany. Tel: (49) 40 34 6184; Fax: (49) 40 35 1117; E-mail: [email protected] Conflict of Interest: Hartmut Porst is a consultant, study investigator, and lecturer for Bayer/GSK, Cilag-Janssen, Lilly/ICOS, Johnson & Johnson, and Pfizer. Claire Gilbert, Suzanne Collins, Xiangning Huang, Tara Symonds, Vera Stecher, and Kyle Hvidsten are employees of Pfizer Inc./Ltd. References
1 Prins J, Blanker MH, Bohnen AM, Thomas S, Bosch JL. Prevalence of erectile dysfunction: A systematic review of population-based studies. Int J Impot Res 2002;14:422–32. 2 Shabsigh R, Perelman MA, Laumann EO, Lockhart DC. Drivers and barriers to seeking treatment for erectile dysfunction: A comparison of six countries. BJU Int 2004;94:1055–65. 3 Wespes E, Amar E, Hatzichristou D, Montorsi F, Pryor J, Vardi Y. Guidelines on erectile dysfunction. Eur Urol 2002;41:1–5. 4 Montague DK, Jarow JP, Broderick GA, Dmochowski RR, Heaton JP, Lue TF, Milbank AJ, Nehra A, Sharlip ID. Chapter 1: The management of erectile dysfunction: An AUA update. J Urol 2005;174:230–9. 5 Lue TF, Giuliano F, Montorsi F, Rosen RC, Andersson KE, Althof S, Christ G, Hatzichristou D, Hirsch M, Kimoto Y, Lewis R, McKenna K, MacMahon C, Morales A, Mulcahy J, PadmaNathan H, Pryor J, de Tejada IS, Shabsigh R, Wagner G. Summary of the recommendations on sexual dysfunctions in men. J Sex Med 2004;1:6–23. 6 NIH Consensus Conference. Impotence. NIH consensus development panel on impotence. JAMA 1993;270:83–90. 7 Rosen R, Althof S, Giuliano F. Research instruments for the diagnosis and treatment of patients with erectile dysfunction. Urology 2006;68(3A suppl):6–16. 8 Rosen RC, Riley A, Wagner G, Osterloh IH, Kirkpatrick J, Mishra A. The International Index of Erectile Function (IIEF): A multidimensional scale for assessment of erectile dysfunction. Urology 1997;49:822–30. 9 Rosen RC, Cappelleri JC, Gendrano N 3rd. The International Index of Erectile Function (IIEF): A state-of-the-science review. Int J Impot Res 2002;14:226–44. 10 Hirsch M, Donatucci C, Glina S, Montague D, Montorsi F, Wyllie M. Standards for clinical trials in male sexual dysfunction: Erectile dysfunction and rapid ejaculation. J Sex Med 2004;1:87–91. 11 Cappelleri JC, Rosen RC, Smith MD, Mishra A, Osterloh IH. Diagnostic evaluation of the erectile function domain of the International Index of Erectile Function. Urology 1999;54:346–51. J Sex Med 2007;4:372–381
380 12 Cappelleri JC, Siegel RL, Osterloh IH, Rosen RC. Relationship between patient self-assessment of erectile function and the erectile function domain of the International Index of Erectile Function. Urology 2000;56:477–81. 13 Rosen RC, Cappelleri JC, Smith MD, Lipsky J, Peña BM. Development and evaluation of an abridged, 5item version of the International Index of Erectile Function (IIEF-5) as a diagnostic tool for erectile dysfunction. Int J Impot Res 1999;11:319–26. 14 de Boer BJ, Bots ML, Lycklama a Nijeholt AA, Moors JPC, Verheij TJ. The prevalence of bother, acceptance, and need for help in men with erectile dysfunction. J Sex Med 2005;2:445–50. 15 Braun M, Wassmer G, Klotz T, Reifenrath B, Mathers M, Engelmann U. Epidemiology of erectile dysfunction: Results of the ‘Cologne Male Survey’. Int J Impot Res 2000;12:305–11. 16 Pfizer Inc. Data on file. New York: Pfizer Inc; 2005. 17 Coleman C, Carabino J, Vergara C, Wang F. Vardenafil: An oral selective phosphodiesterase 5 inhibitor for the treatment of erectile dysfunction. Formulary 2003;38:131–48. 18 Wincze J, Rosen R, Carson C, Korenman S, Niederberger C, Sadovsky R, McLeod L, Thibonnier M, Merchant S. Erection quality scale: Initial scale development and validation. Urology 2004;64:351–6. 19 DiBenedetti DB, Gondek K, Sagnier PP, Kubin M, Marquis P, Keininger D, Fugl-Meyer AR. The treatment satisfaction scale: A multidimensional instrument for the assessment of treatment satisfaction for erectile dysfunction patients and their partners. Eur Urol 2005;48:503–11. 20 Fisher WA, Brock G, Karlin G, Pommerville P, Huang XY, Bangerter K, Herman-Gnjidic Z, Derogatis L, Group VS, Rosen RC. Vardenafil improves erection quality assessed by the novel erection quality scale in the broad population of men with erectile dysfunction. J Sex Med 2006;3(3 suppl):251–2. 21 Lowy M, Collins S, Bloch MT, Gillman M, Lording DW, Sutherland P, Wang H, Stecher V, and the QEQ Study Group. Quality of erection questionnaire correlates: Change in erection quality with erectile function, hardness, and psychosocial measures in men treated with sildenafil for erectile dysfunction. J Sex Med 2007;4:83–92. 22 Althof SE, O’Leary MP, Cappelleri JC, Hvidsten K, Stecher VJ, Glina S, King R, Siegel RL. Sildenafil citrate improves self-esteem, confidence, and relationships in men with erectile dysfunction: Results from an international, multi-center, double-blind, placebo-controlled trial. J Sex Med 2006;3:521–9. 23 Cappelleri JC, Bell SS, Althof SE, Siegel RS, Stecher VJ. Comparison between sildenafil-treated subjects with erectile dysfunction and control subjects on the Self-Esteem And Relationship Questionnaire. J Sex Med 2006;3:274–82. 24 Althof SE, Corty EW, Levine SB, Levine F, Burnett AL, McVary K, Stecher V, Seftel AD. EDITS: J Sex Med 2007;4:372–381
Porst et al. Development of questionnaires for evaluating satisfaction with treatments for erectile dysfunction. Urology 1999;53:793–9. Appendix 1
Quality of Erection Questionnaire The following questions ask about the quality of your erections over the past 4 weeks. Please check the box that best fits your response for each question. In answering these questions, the following definitions apply: Sexual activity includes intercourse, caressing, foreplay, and masturbation. Sexual intercourse is defined as (vaginal) penetration of the partner (you entered your partner). 1. You had erections hard enough for penetration of your partner: Almost always or always More than half the time About half the time Less than half the time Almost never or never 2. Your ability to keep your erection to completion of sexual intercourse was: Very satisfactory Somewhat satisfactory Neither satisfactory nor unsatisfactory Somewhat unsatisfactory Very unsatisfactory 3. The length of time (from when you started sexual activity) until your erection was hard enough to participate in sexual intercourse was: Very satisfactory Somewhat satisfactory Neither satisfactory nor unsatisfactory Somewhat unsatisfactory Very unsatisfactory 4. The length of time you were able to be erect during intercourse was: Very satisfactory Somewhat satisfactory Neither satisfactory nor unsatisfactory Somewhat unsatisfactory Very unsatisfactory 5.
The hardness of your erection was: Very satisfactory Somewhat satisfactory Neither satisfactory nor unsatisfactory Somewhat unsatisfactory Very unsatisfactory
QEQ Development and Validation 6.
381
The overall quality of your erection was: Very satisfactory Somewhat satisfactory Neither satisfactory nor unsatisfactory Somewhat unsatisfactory Very unsatisfactory
Appendix 2
Scoring the QEQ The QEQ is to be evaluated as a total score, which is the sum of responses to all items transformed onto a 0–100 scale. The raw score may be used informally (e.g., in the clinical practice setting). However, Pfizer clinical trial data are reported as transformed scores, and transformed scores are recommended for use in clinical trials. No. items Concepts covered Time to complete Scoring instructions
Lowest possible score Highest possible score Missing data
6 Quality of erections Estimated, 3 minutes Total score, 0–100 (highest = better quality) All items scored as follows: Item 1 Almost always or always 5 More than half the time 4 About half the time 3 Less than half the time 2 Almost never or never 1 Items 2–6 Very satisfactory 5 Somewhat satisfactory 4 Neither satisfactory nor unsatisfactory 3 Somewhat unsatisfactory 2 Very unsatisfactory 1 Sum of all six items is transformed onto 0–100 scale using the following formula: (raw unstandardized score − N items) × 100/(raw score range) If all six items have been completed this would be: (raw unstandardized score − 6) × 100/24 0 100 If four or more items are answered, the standardized score can still be calculated by adjusting the formula accordingly. For example, if four items had been answered, the equation would be: (raw unstandardized score − 4) × 100/16 If there are more than two items missing, a score cannot be calculated.
For more information contact [email protected] © 2005 Pfizer Ltd. All rights reserved
J Sex Med 2007;4:372–381
372
ORIGINAL RESEARCH—OUTCOMES ASSESSMENT Development and Validation of the Quality of Erection Questionnaire Hartmut Porst, MD,* Claire Gilbert, MSc,† Suzanne Collins, BA,‡ Xiangning Huang, PhD,† Tara Symonds, PhD,† Vera Stecher, PhD,‡ and Kyle Hvidsten, MPH‡ *Private Practice of Urology/Andrology, Hamburg, Germany; †Pfizer Ltd., Sandwich, UK; ‡Pfizer Inc, New York, NY, USA DOI: 10.1111/j.1743-6109.2006.00422.x
ABSTRACT
Introduction. There are no psychometrically validated assessment tools designed to solely and specifically evaluate satisfaction with the quality of erections. Aim. To develop and psychometrically analyze the Quality of Erection Questionnaire (QEQ), a new patientreported measure developed to evaluate men’s satisfaction with the quality of their erections. Methods. The questionnaire was developed through in-depth qualitative interviews of men with erectile dysfunction (ED) in the United States and Australia. An exploratory methodology study was conducted on 65 men with ED. Subsequently, the psychometric properties were confirmed in a larger dataset of 558 men with ED from two combined clinical trials. Main Outcome Measures. Identification of potential redundancy or outliers in items (Pearson inter-item correlations); exploratory factor analysis (unrotated and varimax rotated); internal consistency (Cronbach’s alpha); convergent validity (Pearson correlation coefficients between the QEQ total score and domain scores of the International Index of Erectile Function); known-groups validity (ability of the QEQ scores to differentiate between ED severity groups); test–retest reliability (Pearson correlation coefficient). Results. The QEQ demonstrated excellent convergent and known-groups validity. Additional analysis demonstrated high internal consistency (Cronbach’s alpha, 0.92). Item analysis demonstrated a unidimensional structure and suggested that satisfaction with hardness may be the key driver for satisfaction with overall quality of erections (r = 0.8). The smaller exploratory study demonstrated good test–retest reliability (r = 0.82). Conclusions. The QEQ is a six-item, patient-reported measure with a unidimensional structure, which produces a total score that may be transformed to a 0–100 scale. Psychometric analysis confirmed reliability and validity of the QEQ, which solely and specifically evaluates satisfaction of men with the quality of their erections. The QEQ is a potentially useful measure for monitoring and evaluating treatment in those who are bothered by, or concerned about, their erectile function. Porst H, Gilbert C, Collins S, Huang X, Symonds T, Stecher V, and Hvidsten K. Development and validation of the Quality of Erection Questionnaire. J Sex Med 2007;4:372–381. Key Words. Erectile Dysfunction; Sildenafil; Quality of Erection Questionnaire; PDE 5 Inhibitors
Introduction
P
opulation-based data indicate that erectile dysfunction (ED) affects a substantial proportion of men worldwide [1]. In a cross-national survey, the prevalence of men aged 20–75 years J Sex Med 2007;4:372–381
with ED was 19–25% in the United States, Germany, Italy, and the UK, and was 12–13% in France and Spain [2]. The American Urological Association, the European Association of Urology, and the 2nd International Consultation on Sexual Medicine recommend oral phosphodiesterase type © 2006 International Society for Sexual Medicine
373
QEQ Development and Validation 5 inhibitors (PDE5Is; e.g., sildenafil citrate [Viagra®, Pfizer Inc], tadalafil [Cialis®, Lilly/ICOS], and vardenafil [Levitra®, Bayer/GSK]) as first-line oral therapy for ED, unless contraindicated [3–5]. The National Institute of Health Consensus Conference recommended improved “quantitative and qualitative classification of erectile dysfunction” and “standardization of patient assessment and treatment outcome” [6]. In response, several tools have been developed. The International Index of Erectile Function (IIEF), a 15-item patient-reported questionnaire accepted by regulatory agencies and scientific journals [7], is a standard tool for evaluating the efficacy of therapeutic intervention in men with ED in clinical trials [8– 10]. The IIEF Erectile Function domain is used to diagnose and classify severity of ED and to measure treatment effects in clinical research [9,11,12]. The abridged five-item version of the IIEF (IIEF-5), also known as the Sexual Health Inventory for Men (SHIM), was developed to diagnose the presence and severity of ED [13]. The inclusion of satisfaction domains in the IIEF (Intercourse Satisfaction and Overall Satisfaction) and of a single satisfaction question in the IIEF-5/SHIM (Q5: When you attempted sexual intercourse, how often was it satisfactory for you?) reflects the understanding that concern or bother of the patient is important to the evaluation and clinical management of ED. Concern or bother is essential to motivate the patient to seek and maintain treatment. Concern or bother about ED declines with increasing age of the patient [14,15]. In the previously mentioned cross-national survey, lack of concern or bother is inherent in some of the barriers cited most frequently by treatment-naive men as preventing their obtaining treatment, for example, the belief that ED is a normal part of aging (44%) and regarding ED as unimportant (17%) [2]. An international survey of more than 3,500 men with ED established that quality of erections, and specifically the capability to enable hard erections, is the primary attribute sought in a treatment for ED [16]. Although quality of erections is among the concepts included in the IIEF, its Erectile Function domain, and the IIEF-5/SHIM, hardness per se is addressed by only a single item in these questionnaires (Q2: When you had erections with sexual stimulation, how often were your erections hard enough for penetration?) and satisfaction with quality of erections is not addressed at all.
This report documents the development and psychometric validation of the Quality of Erection Questionnaire (QEQ), a patient-reported outcome for solely and specifically evaluating and monitoring satisfaction with the quality of erections in men who are concerned or bothered by their erectile function. Methods
Questionnaire Development The initial conceptual framework and the pool of items were developed through in-depth qualitative interviews of 35 men with ED, aged 18–60 years. The interviews were conducted in the United States (N = 20) and Australia (N = 15). Participants within the United States either were identified through lists of patients involved in clinical trials of ED medication or were respondents to advertisements in local newspapers. Participants within Australia were identified through routine outpatient and general practitioner consultation. All participants gave informed consent. They were required to have experience with some kind of treatment for their ED within the last 6 months and to be currently involved in a sexual relationship. ED was confirmed by a score of 21 or less (indicating at least mild-tomoderate severity) on the IIEF Erectile Function domain. The interviews began with a semi-structured open discussion to obtain perspectives of each individual on factors that relate to the quality of erections. This was followed by a request for feedback on a list of potential items that were suggested by a review of the literature. The semistructured discussions were analyzed to develop patient-derived concepts. Based on analysis of the interviews, a seven-item draft questionnaire was developed. Psychometric Evaluation Studies Exploratory Study. In a small exploratory study, the structure of the draft questionnaire was assessed, test–retest reliability was determined, and preliminary validation data were obtained. This study included 65 men with ED who were recruited during routine clinical consultation. Inclusion criteria were age 18–65 years, being in a sexual relationship, and receipt of treatment for ED for 2 months within the 6 months before enrollment. The draft questionnaire was comJ Sex Med 2007;4:372–381
374 pleted twice, with a minimum 10-day interval between completions. Clinical Trials Study. For the primary analysis of the psychometric properties of the questionnaire, data were obtained from the combined populations of two clinical trials, both assessing the efficacy and safety of an investigational new-generation PDE5I for the treatment of ED. This provided a larger population (N = 558), lending greater strength to the results. These two trials used identical inclusion/exclusion criteria, including recent successful treatment with a PDE5I (a washout period ensured that they were untreated at baseline). Baseline characteristics were compared statistically (Wilcoxon rank sum test for ordinal variables [i.e., age, duration of ED, and scores for the draft questionnaire and the IIEF] and chi-square for categorical variables [i.e., smoking and alcohol use]) to ensure the appropriateness of combining the data into a single, large dataset. The psychometric analyses were based on baseline data from the draft QEQ and IIEF. The trials were performed in compliance with relevant laws and institutional guidelines. Patients provided written informed consent before participation.
Psychometric Analyses Pearson inter-item correlations were used to identify any potential redundancy or outliers in items. Specifically, items that correlated highly (correlation coefficient, r > 0.8; e.g., coherence) were considered further to explore whether both items were needed (redundancy), and items that correlate poorly (r < 0.3) were considered further to examine whether they were covering the same concepts as other questions (outliers). The questionnaire’s underlying structure was evaluated using exploratory factor analysis (unrotated and varimax rotated); a loading of >0.5 delineated unidimensionality of items. Internal consistency was assessed by calculating Cronbach’s alpha. Convergent validity was evaluated by determining parametric correlation (Pearson correlation coefficients) between the QEQ total score and the IIEF domain scores. Known-groups validity was also evaluated by determining the ability of the QEQ scores (total and items) to differentiate between ED severity groups as defined by scores on the Erectile Function domain of the IIEF: 26– 30 (no ED), 22–25 (mild ED), 17–21 (mild-tomoderate ED), 11–16 (moderate ED), and 6–10 (severe ED). Because the QEQ scores were not J Sex Med 2007;4:372–381
Porst et al. normally distributed, nonparametric one-way analysis of variance was applied. Test–retest reliability was assessed across a 2week interval using Pearson correlation coefficient (r) ≥0.8. Because the protocols of the clinical trials did not provide a period of no expected change in ED, test–retest (unlike the other results reported herein) was based on data from the exploratory study. Results
The exploratory study assessing test–retest reliability included 65 men aged 22–75 years (mean ± SD, 54.7 ± 10.6 years) with ED duration of 2 months to 20 years (6.1 ± 4.4 years). ED medication had first been used from 1 month to 18 years previously (4.3 ± 3.8 years), 57 of 63 respondents (91%) indicated that they were currently using the prescribed medication, and usage was from 0 to 15 times per month (5.1 ± 3.4). The most commonly prescribed medications for ED were sildenafil (N = 54) and tadalafil (N = 44). The clinical trials enrolled 594 men aged 23– 65 years (53.8 ± 8.1 years) with ED duration of 0.5–35 years (6.2 ± 4.9 years), 558 of whom completed the draft questionnaire and composed the main study population. Of these, erectile function was categorized according to the IIEF Erectile Function domain in 546, including 214 (39%) with severe ED, 174 (32%) with moderate ED, 115 (21%) with mild-to-moderate ED, 36 (7%) with mild ED, and seven (1%) with no ED (who were included in the clinical trials based on a clinical diagnosis of ED). The mean IIEF Erectile Function domain score was 12.6 (out of 30). Between the two clinical trial populations that composed the main study population (N = 223 and N = 335, respectively), statistically significant differences were found in reported alcohol use (62% [119/191] vs. 76% [208/272] of respondents), the mean duration of ED (5.3 [range 0.5– 31] vs. 6.9 [range 0.7–35] years), and mean IIEF Orgasmic Function domain scores (6.0 vs. 5.5 out of 10) at baseline. These differences were not expected to affect the relationships between the QEQ and the other endpoints.
Quality of Completion There was a low rate of missing items—only eight of 558 had at least one missing item. There were no observed floor or ceiling effects, defined as an endorsement >50% for any response option. The highest endorsement frequencies were at the
375
QEQ Development and Validation Table 1
Parametric correlations between item scores on the draft questionnaire Pearson’s correlation coefficient (r value), N = 550* Frequency items
Satisfaction items
Draft questionnaire items†
Hard enough
Loss of erection
Keep erection
Time to hard enough
Time erect
Hardness
Overall quality
Hard enough Loss of erection Keep erection Time to hard enough Time erect Hardness Overall quality
1.00 0.38 0.58 0.63 0.55 0.62 0.61
— 1.00 0.44 0.26 0.40 0.36 0.40
— — 1.00 0.57 0.74 0.65 0.72
— — — 1.00 0.65 0.69 0.64
— — — — 1.00 0.71 0.75
— — — — — 1.00 0.82
— — — — — — 1.00
*Bold indicates outcomes that are highly correlated (>0.8) or poorly correlated (<0.3). † Draft questionnaire items were as follows: Frequency that: You had erections hard enough for penetration of your partner. You experienced the loss of your erection within a few minutes of penetration of your partner. Satisfaction with: Your ability to keep your erection to completion of sexual intercourse. The length of time (from when you started sexual activity) until your erection was hard enough to participate in sexual intercourse. The length of time you were able to be erect during intercourse. The hardness of your erection. The overall quality of your erection.
lower end of the score scale, indicating lower erection quality and satisfaction. This was expected from a population of men whose ED is untreated. There were no endorsements at the higher end of the score scale, suggesting room to capture any improvements in erection quality and in satisfaction.
Questionnaire Structure Inter-item correlations were relatively low between the “loss of erection” item (i.e., the frequency that “You experienced the loss of your erection within a few minutes of penetration of your partner”) and all other items (Table 1). This suggested that the “loss of erection” item may be an outlier, which needed to be evaluated further to determine its relationship to the other items and to the overall questionnaire structure. The strong correlation (r > 0.8) between the items questioning satisfaction with “The hardness of your erection” and “The overall quality of your erection” suggested potential redundancy. It could be that hardness is the key driver for overall satisfaction. All other items correlated moderately with each other (r = 0.55–0.75), with no obvious clusters of highly correlated items, suggesting a lack of redundancy and that the questionnaire is unidimensional. Factor analyses were used to identify the most appropriate factor structure for the questionnaire (Table 2). The results, quantified by eigenvalues, suggested a one-factor solution in which all seven items measure a single factor, accounting for 65%
of the variance. All other factors accounted for 12% or less of the variance. All items loaded onto the factor with a loading >0.5, which supports the unidimensionality of the questionnaire. However, the “loss of erection” item had a loading of 0.53, which was markedly lower than those of the other items (0.78–0.89). This suggests that the “loss of erection” item may not be a strong item in the overall factor structure. Further support for this conclusion came from this item’s low communality estimate (h2 = 0.28) compared with those of the other items (h2 = 0.61–0.79). This suggested that the “loss of erection” item (unlike the other items) is not strongly influenced by the one factor identified. Thus, the results of the item and factor analysis gave strong support for the questionnaire evaluating a unidimensional construct, with the removal of the “loss of erection” item. The resultant sixitem questionnaire, hereafter referred to as the QEQ, was scored by a single total score, calculated as the sum of the scores of all six items transformed onto a 0–100 scale (higher score = better quality).1 On this scale, the mean QEQ total score for the main study population was 25.6 ± 23.4. The transformed QEQ total score correlated positively with each item, with
1
The raw score may be used informally (e.g., in the clinical practice setting). However, Pfizer clinical trial data are reported as transformed scores, and transformed scores are recommended for use in clinical trials.
J Sex Med 2007;4:372–381
376
Porst et al.
Table 2 Factor analysis of the draft questionnaire: rotated and unrotated to identify the most appropriate factor structure, item loading to support unidimensionality, and communality estimates of the items Eigenvalues of the correlation matrix, N = 550 Eigenvalue Difference
Proportion
Cumulative
4.56 0.81 0.52 0.37 0.33 0.23 0.18
0.65 0.12 0.07 0.05 0.05 0.03 0.03
0.65 0.77 0.85 0.90 0.94 0.98 1.0
3.75 0.29 0.16 0.04 0.11 0.05 —
Factor pattern and communality estimates, N = 550
Factor pattern: item loading onto factor 1 (rotated and unrotated) Final communality estimates (h 2): factor 1 (rotated and unrotated): Total = 4.6
Frequency items*
Satisfaction items*
Hard enough
Loss of erection
Keep erection
Time to hard enough
Time erect
Hardness
Overall quality
0.78
0.53
0.85
0.80
0.87
0.88
0.89
0.61
0.28
0.72
0.64
0.75
0.77
0.79
*Draft questionnaire items were as follows: Frequency that: You had erections hard enough for penetration of your partner. You experienced the loss of your erection within a few minutes of penetration of your partner. Satisfaction with: Your ability to keep your erection to completion of sexual intercourse. The length of time (from when you started sexual activity) until your erection was hard enough to participate in sexual intercourse. The length of time you were able to be erect during intercourse. The hardness of your erection. The overall quality of your erection.
Spearman correlation coefficients ranging from 0.82 to 0.88 (Table 3). Given the high factor loading of the item about satisfaction with “the overall quality of your erection,” the factor analysis was rerun on the final QEQ without this item to ensure that this one item was not itself determining the single factor structure. The factor structure did not change with this reanalysis. The results, quantified by eigenvalues, suggested a one-factor solution in Table 3 Spearman’s nonparametric correlations between the QEQ total score (representing the unidimensionality of the questionnaire) and the scores for each item QEQ items†
Correlations (r values)*
Item Item Item Item Item Item
0.82 0.83 0.85 0.86 0.88 0.87
1: hard enough 2: keep erection 3: time to hard enough 4: time erect 5: hardness 6: overall quality
*Significant for all correlation coefficients, with P < 0.0001 (N = 550). † QEQ items were as follows: Frequency that: Item 1: You had erections hard enough for penetration of your partner. Satisfaction with: Item 2: Your ability to keep your erection to completion of sexual intercourse. Item 3: The length of time (from when you started sexual activity) until your erection was hard enough to participate in sexual intercourse. Item 4: The length of time you were able to be erect during intercourse. Item 5: The hardness of your erection. Item 6: The overall quality of your erection. QEQ = Quality of Erection Questionnaire.
J Sex Med 2007;4:372–381
which the remaining items measured a single factor, accounting for 71% of the variance. All other factors accounted for 10% or less of the variance. All items loaded onto the factor with a loading >0.79, which supports the unidimensionality of the questionnaire. A unidimensional structure was therefore accepted for the QEQ, using all six items of the final measure.
Reliability Reliability was demonstrated by internal consistency and by test–retest analysis. The interval between the first and second completion of the questionnaire for test–retest ranged from 10 to 50 days (mean ± SD, 19 ± 7.7 days). Fifty-four pairs of questionnaires were available for analysis by Pearson correlation coefficient. The QEQ had excellent internal consistency (Cronbach’s alpha = 0.92) and demonstrated good retest reliability (correlation coefficient, 0.82). Validity Because convergent validity demonstrates that endpoints are closely related (although not identical), it was expected that the strongest convergent relationships of the QEQ total score would be with the IIEF Overall Satisfaction domain and Erectile Function domain. The results confirmed
377
QEQ Development and Validation Table 4
Correlations between transformed QEQ scores and IIEF domain scores Pearson’s correlation coefficient (r value)* IIEF domain
QEQ†
Erectile function
Orgasmic function
Sexual desire
Intercourse satisfaction
Overall satisfaction
Total score Item 1. hard enough Item 2. keep erection Item 3. time to hard enough Item 4. time erect Item 5. hardness Item 6. overall quality
0.67 0.73 0.55 0.55 0.52 0.54 0.53
0.37 0.37 0.30 0.31 0.27 0.29 0.29
0.13 0.11 0.08 0.17 0.08 0.13 0.10
0.47 0.51 0.36 0.40 0.36 0.38 0.36
0.58 0.50 0.49 0.46 0.48 0.50 0.51
*P < 0.0001 except for correlations between the Sexual Desire domain and the QEQ total score, item 1, and item 5 (P < 0.01); item 6 (P < 0.05); and items 2 and 4 (P > 0.05, not statistically significant at the 0.05 level). † QEQ items were as follows: Frequency that: Item 1: You had erections hard enough for penetration of your partner. Satisfaction with: Item 2: Your ability to keep your erection to completion of sexual intercourse. Item 3: The length of time (from when you started sexual activity) until your erection was hard enough to participate in sexual intercourse. Item 4: The length of time you were able to be erect during intercourse. Item 5: The hardness of your erection. Item 6: The overall quality of your erection. QEQ = Quality of Erection Questionnaire; IIEF = International Index of Erectile Function.
this expectation, demonstrating correlations in the range expected for convergent validity for these domains, as well as for the Intercourse Satisfaction domain (r = 0.47–0.67) (Table 4). In the known-groups validity analysis, mean ± SD QEQ total scores followed the expected direction, being highest in the men categorized as having no ED and decreasing with each subsequent category (Table 5). The same relationship was seen for each individual QEQ item (Figure 1). Statistical analysis showed that these differences were significant (P < 0.0001 for an F value of 109.64 > F0.0001(4,541)). These results suggest that the QEQ is able to differentiate between all groups categorized by ED severity (P < 0.0001, with the exception of no ED vs. mild ED [P = 0.0031]), thereby demonstrating good known-groups validity.
Table 5 Nonparametric post-hoc analysis: difference in QEQ total score by ED severity QEQ total score ED severity*
Mean ± SD
Range
No ED (N = 7) Mild ED (N = 36) Mild-to-moderate ED (N = 115) Moderate ED (N = 174) Severe ED (N = 214)
79.8 ± 11.1 59.7 ± 16.4 39.9 ± 14.9 25.5 ± 17.5 10.5 ± 19.1
62.5–91.7 33.3–100 4.2–79.2 0–87.5 0–100
*ED severity was categorized by score on the Erectile Function domain of the International Index of Erectile Function. ED = erectile dysfunction; QEQ = Quality of Erection Questionnaire.
Discussion
The results show that the QEQ has good psychometric properties. The original seven items were reduced to a six-item unidimensional questionnaire. All items demonstrated good internal consistency with a strong factor structure and good test–retest reliability. The detailed analyses of the items and factor structure of the measure suggest that hardness is the key driver for overall satisfaction with the quality of erections, confirming the results of the international survey of more than 3,500 men with ED [16]. The concept that hardness is a key driver of overall satisfaction with the quality of erections is also confirmed by the qualitative work supporting the development of the questionnaire, in which some men reported overlap between the two items. Although this and the high level of correlation found between these items could suggest potential item redundancy, other men indicated as part of the initial qualitative work that overall quality for them reflected not only hardness but also other factors such as the additional QEQ concepts and emotional elements. The overall quality of erections item gives the respondent the chance to capture all the elements that are important to them in their overall perception of quality of their erections, whether that be primarily a matter of hardness for some or an amalgamation of several factors for others. Convergent and known-groups validity was demonstrated against the IIEF, a documented J Sex Med 2007;4:372–381
378
Porst et al. No ED (N=7) Mild ED (N=36) Mild-to-Moderate ED (N=116) Moderate ED (N=176) Severe ED (N=217–218)
QEQ Score by IIEF ED Severity Group
5
4
3
2
1
0 Item 1: Hard Enough
Item 2: Keep Erection
Item 3: Time to Hard Enough
Item 4: Time Erect
Item 5: Hardness
standard for evaluating ED. With the exception of the inability to differentiate between the two least impaired categories (“no ED” and “mild ED”), the QEQ was able to differentiate between all severity groups of ED categorized by IIEF Erectile Function domain scores. The inability to differentiate between “no ED” and “mild ED” may be the result of the small sample size of these groups (N = 36 and N = 7, respectively). Although the ability of the QEQ to differentiate between functional men and men with compromised erectile function warrants further research, the clear differentiation among the ED severity groups in the current study suggests that differentiation from functional men is likely given sufficient sample size. Although the IIEF Erectile Function domain and the QEQ total score are highly correlated (r = 0.67), the IIEF Erectile Function domain was designed to assess erectile function. The QEQ provides a further level of information by assessing satisfaction with the quality of the erections that were attained and maintained. The QEQ addresses the unmet need for a measure to solely and specifically assess the key concept of satisfaction with quality of erections from the individual male/ patient perspective. Quality of erections is composed of several aspects, such as the degree of hardness, the functionality of hardness, the time to onset of hardness, and the duration of hardness. Most previous methods for assessing quality of J Sex Med 2007;4:372–381
Item 6: Overall Quality
Figure 1 Known-groups discrimination: QEQ item score according to ED severity. ED severity was categorized by score on the Erectile Function domain of the IIEF. Item 1: (frequency that) You had erections hard enough for penetration of your partner; item 2: (satisfaction with) Your ability to keep your erection to completion of sexual intercourse; item 3: (satisfaction with) The length of time (from when you started sexual activity) until your erection was hard enough to participate in sexual intercourse; item 4: (satisfaction with) The length of time you were able to be erect during intercourse; item 5: (satisfaction with) The hardness of your erection; item 6: (satisfaction with) The overall quality of your erection. ED = erectile dysfunction; IIEF = International Index of Erectile Function; QEQ = Quality of Erection Questionnaire.
erections only addressed individual aspects: the unvalidated Erection Hardness Score, which allows a man to subjectively classify hardness on a simple 4-point scale at the time of erection; individual unvalidated items from validated questionnaires (e.g., IIEF Q2 [frequency of erection hard enough for penetration]); and individual questions with dichotomous (yes/no) response options, such as a global efficacy question or the questions of the Sexual Encounter Profile (“Were you able to insert your penis into your partner’s vagina?” and “Did your erection last long enough for you to have successful intercourse?”) [17]. In contrast, two recently developed questionnaires are much broader [18–20], The 15-item Erection Quality Scale (EQS) addresses several aspects of erection quality (six items) and also qualifies confidence in erection quality (three items) and satisfaction with erection quality (four items); the other two items address pleasurable feelings [18,20]. The 13-item Treatment Satisfaction Scale (TSS) addresses erection quality (one item) and also qualifies confidence in erection quality (two items) and satisfaction with erection quality (three items); the seven other items address pleasurable feelings and ED treatment [19]. Although both the EQS and the TSS are valuable tools, only the QEQ focuses solely and specifically on satisfaction with the quality of erections, which is a key quality-of-life measure from the individual male/patient perspective because lack of satisfaction is a prerequisite to
379
QEQ Development and Validation concern or bother, the drivers of treatment-seeking behavior. The quality of the data obtained suggests that the QEQ is acceptable to the ED population. This is to be expected because the items were derived from qualitative work with this population. Further analyses are being or have been conducted on the QEQ, including responsiveness/sensitivity to change, confirming the validity of the QEQ. A recent open-label, noncomparative, multicenter trial of sildenafil (50 or 100 mg as needed for 10 weeks) in men with ED who were in a stable, sexual relationship for at least 6 months and had limited previous PDE5I use (≤6 doses ever and no doses more recently than the previous 4 weeks) found that the mean ± SD transformed QEQ total score tripled from 22.0 ± 21.1 to 69.9 ± 35.9 (P < 0.0001) and the change correlated positively with change scores on the IIEF domains (r = 0.29– 0.86), change scores on measures of emotional well-being from the Self-Esteem And Relationship questionnaire (SEAR, r = 0.37–0.78), and the end-of-treatment score on the Erectile Dysfunction Inventory of Treatment Satisfaction (EDITS, r = 0.71) [21]. The SEAR questionnaire is a validated 14-item questionnaire consisting of a Sexual Relationship domain, a Confidence domain (i.e., Self-Esteem and Overall Relationship subscales), and an Overall score; it has been shown in controlled clinical trials to be responsive to effective treatment of ED with sildenafil [22,23]. The EDITS is an 11-item validated questionnaire that evaluates satisfaction with treatment for ED [24]. Thus, a relationship has been established between the broader psychosocial status of men with ED and satisfaction with the quality of their erections. The QEQ is a brief (six-item) and easy-to-use measure that solely and specifically evaluates satisfaction with the quality of erections from the patient’s perspective. This perspective enables the evaluation of treatment against the perceived need of the patient, and thus the QEQ is a potentially useful measure for use in clinical trials research or in the clinical practice setting. Details of the QEQ are given in the Appendixes. Acknowledgments
This study was sponsored by Pfizer Inc. Editorial support was provided by Deborah M. Campoli-Richards, RPh, of Complete Healthcare Communications, Inc., and was funded by Pfizer Inc. Correspondence Private Practice
Author: Hartmut Porst, MD, of Urology/Andrology, Neuer
Jungfernstieg 6a, Hamburg, 20354. Germany. Tel: (49) 40 34 6184; Fax: (49) 40 35 1117; E-mail: [email protected] Conflict of Interest: Hartmut Porst is a consultant, study investigator, and lecturer for Bayer/GSK, Cilag-Janssen, Lilly/ICOS, Johnson & Johnson, and Pfizer. Claire Gilbert, Suzanne Collins, Xiangning Huang, Tara Symonds, Vera Stecher, and Kyle Hvidsten are employees of Pfizer Inc./Ltd. References
1 Prins J, Blanker MH, Bohnen AM, Thomas S, Bosch JL. Prevalence of erectile dysfunction: A systematic review of population-based studies. Int J Impot Res 2002;14:422–32. 2 Shabsigh R, Perelman MA, Laumann EO, Lockhart DC. Drivers and barriers to seeking treatment for erectile dysfunction: A comparison of six countries. BJU Int 2004;94:1055–65. 3 Wespes E, Amar E, Hatzichristou D, Montorsi F, Pryor J, Vardi Y. Guidelines on erectile dysfunction. Eur Urol 2002;41:1–5. 4 Montague DK, Jarow JP, Broderick GA, Dmochowski RR, Heaton JP, Lue TF, Milbank AJ, Nehra A, Sharlip ID. Chapter 1: The management of erectile dysfunction: An AUA update. J Urol 2005;174:230–9. 5 Lue TF, Giuliano F, Montorsi F, Rosen RC, Andersson KE, Althof S, Christ G, Hatzichristou D, Hirsch M, Kimoto Y, Lewis R, McKenna K, MacMahon C, Morales A, Mulcahy J, PadmaNathan H, Pryor J, de Tejada IS, Shabsigh R, Wagner G. Summary of the recommendations on sexual dysfunctions in men. J Sex Med 2004;1:6–23. 6 NIH Consensus Conference. Impotence. NIH consensus development panel on impotence. JAMA 1993;270:83–90. 7 Rosen R, Althof S, Giuliano F. Research instruments for the diagnosis and treatment of patients with erectile dysfunction. Urology 2006;68(3A suppl):6–16. 8 Rosen RC, Riley A, Wagner G, Osterloh IH, Kirkpatrick J, Mishra A. The International Index of Erectile Function (IIEF): A multidimensional scale for assessment of erectile dysfunction. Urology 1997;49:822–30. 9 Rosen RC, Cappelleri JC, Gendrano N 3rd. The International Index of Erectile Function (IIEF): A state-of-the-science review. Int J Impot Res 2002;14:226–44. 10 Hirsch M, Donatucci C, Glina S, Montague D, Montorsi F, Wyllie M. Standards for clinical trials in male sexual dysfunction: Erectile dysfunction and rapid ejaculation. J Sex Med 2004;1:87–91. 11 Cappelleri JC, Rosen RC, Smith MD, Mishra A, Osterloh IH. Diagnostic evaluation of the erectile function domain of the International Index of Erectile Function. Urology 1999;54:346–51. J Sex Med 2007;4:372–381
380 12 Cappelleri JC, Siegel RL, Osterloh IH, Rosen RC. Relationship between patient self-assessment of erectile function and the erectile function domain of the International Index of Erectile Function. Urology 2000;56:477–81. 13 Rosen RC, Cappelleri JC, Smith MD, Lipsky J, Peña BM. Development and evaluation of an abridged, 5item version of the International Index of Erectile Function (IIEF-5) as a diagnostic tool for erectile dysfunction. Int J Impot Res 1999;11:319–26. 14 de Boer BJ, Bots ML, Lycklama a Nijeholt AA, Moors JPC, Verheij TJ. The prevalence of bother, acceptance, and need for help in men with erectile dysfunction. J Sex Med 2005;2:445–50. 15 Braun M, Wassmer G, Klotz T, Reifenrath B, Mathers M, Engelmann U. Epidemiology of erectile dysfunction: Results of the ‘Cologne Male Survey’. Int J Impot Res 2000;12:305–11. 16 Pfizer Inc. Data on file. New York: Pfizer Inc; 2005. 17 Coleman C, Carabino J, Vergara C, Wang F. Vardenafil: An oral selective phosphodiesterase 5 inhibitor for the treatment of erectile dysfunction. Formulary 2003;38:131–48. 18 Wincze J, Rosen R, Carson C, Korenman S, Niederberger C, Sadovsky R, McLeod L, Thibonnier M, Merchant S. Erection quality scale: Initial scale development and validation. Urology 2004;64:351–6. 19 DiBenedetti DB, Gondek K, Sagnier PP, Kubin M, Marquis P, Keininger D, Fugl-Meyer AR. The treatment satisfaction scale: A multidimensional instrument for the assessment of treatment satisfaction for erectile dysfunction patients and their partners. Eur Urol 2005;48:503–11. 20 Fisher WA, Brock G, Karlin G, Pommerville P, Huang XY, Bangerter K, Herman-Gnjidic Z, Derogatis L, Group VS, Rosen RC. Vardenafil improves erection quality assessed by the novel erection quality scale in the broad population of men with erectile dysfunction. J Sex Med 2006;3(3 suppl):251–2. 21 Lowy M, Collins S, Bloch MT, Gillman M, Lording DW, Sutherland P, Wang H, Stecher V, and the QEQ Study Group. Quality of erection questionnaire correlates: Change in erection quality with erectile function, hardness, and psychosocial measures in men treated with sildenafil for erectile dysfunction. J Sex Med 2007;4:83–92. 22 Althof SE, O’Leary MP, Cappelleri JC, Hvidsten K, Stecher VJ, Glina S, King R, Siegel RL. Sildenafil citrate improves self-esteem, confidence, and relationships in men with erectile dysfunction: Results from an international, multi-center, double-blind, placebo-controlled trial. J Sex Med 2006;3:521–9. 23 Cappelleri JC, Bell SS, Althof SE, Siegel RS, Stecher VJ. Comparison between sildenafil-treated subjects with erectile dysfunction and control subjects on the Self-Esteem And Relationship Questionnaire. J Sex Med 2006;3:274–82. 24 Althof SE, Corty EW, Levine SB, Levine F, Burnett AL, McVary K, Stecher V, Seftel AD. EDITS: J Sex Med 2007;4:372–381
Porst et al. Development of questionnaires for evaluating satisfaction with treatments for erectile dysfunction. Urology 1999;53:793–9. Appendix 1
Quality of Erection Questionnaire The following questions ask about the quality of your erections over the past 4 weeks. Please check the box that best fits your response for each question. In answering these questions, the following definitions apply: Sexual activity includes intercourse, caressing, foreplay, and masturbation. Sexual intercourse is defined as (vaginal) penetration of the partner (you entered your partner). 1. You had erections hard enough for penetration of your partner: Almost always or always More than half the time About half the time Less than half the time Almost never or never 2. Your ability to keep your erection to completion of sexual intercourse was: Very satisfactory Somewhat satisfactory Neither satisfactory nor unsatisfactory Somewhat unsatisfactory Very unsatisfactory 3. The length of time (from when you started sexual activity) until your erection was hard enough to participate in sexual intercourse was: Very satisfactory Somewhat satisfactory Neither satisfactory nor unsatisfactory Somewhat unsatisfactory Very unsatisfactory 4. The length of time you were able to be erect during intercourse was: Very satisfactory Somewhat satisfactory Neither satisfactory nor unsatisfactory Somewhat unsatisfactory Very unsatisfactory 5.
The hardness of your erection was: Very satisfactory Somewhat satisfactory Neither satisfactory nor unsatisfactory Somewhat unsatisfactory Very unsatisfactory
QEQ Development and Validation 6.
381
The overall quality of your erection was: Very satisfactory Somewhat satisfactory Neither satisfactory nor unsatisfactory Somewhat unsatisfactory Very unsatisfactory
Appendix 2
Scoring the QEQ The QEQ is to be evaluated as a total score, which is the sum of responses to all items transformed onto a 0–100 scale. The raw score may be used informally (e.g., in the clinical practice setting). However, Pfizer clinical trial data are reported as transformed scores, and transformed scores are recommended for use in clinical trials. No. items Concepts covered Time to complete Scoring instructions
Lowest possible score Highest possible score Missing data
6 Quality of erections Estimated, 3 minutes Total score, 0–100 (highest = better quality) All items scored as follows: Item 1 Almost always or always 5 More than half the time 4 About half the time 3 Less than half the time 2 Almost never or never 1 Items 2–6 Very satisfactory 5 Somewhat satisfactory 4 Neither satisfactory nor unsatisfactory 3 Somewhat unsatisfactory 2 Very unsatisfactory 1 Sum of all six items is transformed onto 0–100 scale using the following formula: (raw unstandardized score − N items) × 100/(raw score range) If all six items have been completed this would be: (raw unstandardized score − 6) × 100/24 0 100 If four or more items are answered, the standardized score can still be calculated by adjusting the formula accordingly. For example, if four items had been answered, the equation would be: (raw unstandardized score − 4) × 100/16 If there are more than two items missing, a score cannot be calculated.
For more information contact [email protected] © 2005 Pfizer Ltd. All rights reserved
J Sex Med 2007;4:372–381